
==== Front
Mol Ther
Mol Ther
Molecular Therapy
1525-0016
1525-0024
American Society of Gene & Cell Therapy

S1525-0016(23)00261-7
10.1016/j.ymthe.2023.05.005
Original Article
Rescue of auditory function by a single administration of AAV-TMPRSS3 gene therapy in aged mice of human recessive deafness DFNB8
Du Wan 1237
Ergin Volkan 1237
Loeb Corena 123
Huang Mingqian 123
Silver Stewart 123
Armstrong Ariel Miura 123
Huang Zaohua 4
Gurumurthy Channabasavaiah B. 5
Staecker Hinrich 6
Liu Xuezhong x.liu1@med.miami.edu
47∗
Chen Zheng-Yi zheng-yi_chen@meei.harvard.edu
1237∗∗
1 Department of Otolaryngology-Head and Neck Surgery, Graduate Program in Speech and Hearing Bioscience and Technology, Harvard Medical School, Boston, MA 02115, USA
2 Department of Otolaryngology-Head and Neck Surgery, Graduate Program in Neuroscience, Harvard Medical School, Boston, MA 02115, USA
3 Eaton-Peabody Laboratories, Massachusetts Eye and Ear, Boston, MA 02114, USA
4 Department of Otolaryngology, University of Miami Miller School of Medicine, Miami, FL 33136, USA
5 Mouse Genome Engineering Core Facility, University of Nebraska Medical Center, Omaha, NE 68198, USA
6 Kansas University Center for Hearing and Balance Disorders, Kansas City, KS 66160, USA
∗ Corresponding author: Xuezhong Liu, Department of Otolaryngology, University of Miami Miller School of Medicine, 1120 NW 14th St. Miami, FL 33136. x.liu1@med.miami.edu
∗∗ Corresponding author: Zheng-Yi Chen, Eaton-Peabody Laboratories, Massachusetts Eye & Ear, 243 Charles St, Boston 02114. zheng-yi_chen@meei.harvard.edu
7 These authors contributed equally

06 9 2023
26 5 2023
31 9 27962810
8 12 2022
4 5 2023
© 2023 The Author(s)
2023
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Patients with mutations in the TMPRSS3 gene suffer from recessive deafness DFNB8/DFNB10. For these patients, cochlear implantation is the only treatment option. Poor cochlear implantation outcomes are seen in some patients. To develop biological treatment for TMPRSS3 patients, we generated a knockin mouse model with a frequent human DFNB8 TMPRSS3 mutation. The Tmprss3A306T/A306T homozygous mice display delayed onset progressive hearing loss similar to human DFNB8 patients. Using AAV2 as a vector to carry a human TMPRSS3 gene, AAV2-hTMPRSS3 injection in the adult knockin mouse inner ear results in TMPRSS3 expression in the hair cells and the spiral ganglion neurons. A single AAV2-hTMPRSS3 injection in Tmprss3A306T/A306T mice of an average age of 18.5 months leads to sustained rescue of the auditory function to a level similar to wild-type mice. AAV2-hTMPRSS3 delivery rescues the hair cells and the spiral ganglions neurons. This study demonstrates successful gene therapy in an aged mouse model of human genetic deafness. It lays the foundation to develop AAV2-hTMPRSS3 gene therapy to treat DFNB8 patients, as a standalone therapy or in combination with cochlear implantation.

Graphical abstract

Chen and colleagues used AAV-mediated gene therapy for sustained hearing rescue in a mouse model of human genetic hearing loss DFNB8 due to a mutation in the TMPRSS3 gene. This study lays the foundation for developing gene therapy to treat DFNB8 patients as a standalone therapy or in combination with cochlear implantation.

Keywords

gene therapy
genetic hearing loss
DFNB8
DFNB10
recessive
TMPRSS3
AAV
aged mouse
hair cells
spiral ganglion neurons
==== Body
pmcIntroduction

Hearing loss (HL) is one of the most common sensory deficit disorders that affect about 466 million people.1 Expected to afflict 1 in 10 individuals by 2050, HL poses a social and emotional toll, and a growing worldwide annual economic burden.1 HL has been linked to increased instances of social isolation and higher risk for dementia and depression.2,3 Genetic HL affects 1 in 500 newborns. There are no treatments available to reverse or prevent genetic deafness.4,5 Currently hearing aids and cochlear implantation (CI) are the only treatment options, and these require residual hearing function or the ability to stimulate the cochlear nerve, respectively. While genetic HL patients can benefit from CI, those patients with ganglion deficits will be left without any treatment option.

Rapid progress has been made in the understanding of the genetic etiology of human HL.6 Genetic testing and diagnosis of HL provide essential information for further genetic therapies.7,8 Adeno-associated virus (AAV), a non-replicative viral vector with low immunogenicity and little ototoxicity, is one of the most promising gene therapy tools for transducing broad cellular tropism.9,10 Gene therapies including gene editing, which can replace, edit, or silence genes, offer a promising avenue for treatment. Among them, gene replacement is appropriate for treating recessive monogenic disorders and has achieved the most clinical success, e.g., Luxturna11 for Leber’s congenital amourosis (an inherited retinal disease) and Zolgensma12 for spinal muscular atrophy. As the inner ear is an isolated organ that can be accessed safely by local injection, a number of gene replacement and overexpression studies targeting HL have been conducted successfully, resulting in hearing rescue and the survival of hair cells (HC) or spiral ganglion neurons (SGN).13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35

However, with the exception of Otof,25 all the gene therapy studies were performed in neonatal animals, which raises the question of the suitability of the approach in the fully mature adult inner ear.

TMPRSS3 protein, a type II transmembrane serine protease, is necessary for normal hearing in mammals. TMPRSS3 gene mutations account for approximately 12%–13% of HL families that are negative for common genetic mutations.36,37 Two different phenotypes were present in individuals with TMPRSS3 mutations: prelingual (DFNB10, OMIM 605511) and the delayed onset and postlingual (DFNB8, OMIM 601072).38 A previous study has shown TMPRSS3 as a necessary permissive factor for cochlear hair cell activation and survival upon the onset of hearing loss.39 Mice with truncated Tmprss3 protease domains of mutation Tmprss3Y260X showed congenital HL, and rapid loss of HC and progressive loss of SGN.39

To develop a mouse model with a TMPRSS3 mutation suitable for gene therapy intervention, we constructed a DFNB8 mouse model with a knockin (KI) of a human TMPRSS3 mutation (c.916G>A, Ala306→Thr), which causes adulthood onset and progressive recessive HL. We show by injection in aged mice, a human TMPRSS3 gene carried by an AAV2 is re-expressed in the HC and the modiolus region. TMPRSS3 inner ear delivery rescues hearing in aged KI mice, concomitant with the survival of HC and SGN.

Results

Generation of Tmprss3 p.A306T KI mouse model

The TMPRSS3 gene has 13 exons and encodes a protein that consists of 453 amino acids (aa), containing four domains (Figure 1A). A human TMPRSS3 mutation in Ala306 identified in HL patients is associated with DFNB8/10.40,41,42,43,44 Ala306 is highly conserved across species from zebrafish to humans (Figure 1B). We chose to use the inbred strain of CBA/CaJ to create a KI mouse model as CBA/CaJ does not suffer from age-related hearing loss (ARHL), as in the strain of C57BL/6J, and will allow us to analyze the rescue effect over an extended period of time.45 We used CRISPR-Cas9 technology46,47,48,49 to create a KI mutation c.916G>A, which changed Ala to Thr at amino acid position 306 (p.A306T). In brief, to introduce mutations of Tmprss3 c.916G>A;918 C>T, i.e., p.Ala306Thr, a ribonucleoprotein complex of Cas9 protein and UGGCUCGGACAGCUUCAUGA guide RNA with the donor oligonucleotide of ACAGCAAGTACAAGCCAAAGCGGCTGGGCAATGACATAACTC.Figure 1 Generation of Tmprss3 p.A306T KI mouse model with late-onset progressive hearing loss

(A) Summary of human TMPRSS3 protein domains and mutations. (B) Conservation of the 306 alanine from human to zebrafish species. (C) Sanger sequencing of TMPRSS3 c.916G>A;c.918C>T mutant mice. (D–G) ABR thresholds in Tmprss3A306T/A306T homozygous mice ears (red) compared with wild-type ears (blue) at 1.5 months (D), 10.5 months (E), 16.5 months (F), and 22.5 months (G), respectively. Significant hearing loss by the elevated ABR thresholds was seen at 16.5 months, which became more severe at 22.5 months. Values and error bars reflect mean ± SEM.

TCATGAAGCTTTCCGAGCCACTCACCTTTGACGAGACCATCCAGCC were injected into CBA/CaJ zygotes (Jackson Laboratory, stock no. 000656). The microinjected zygotes were then transferred into pseudo-pregnant females.49 The target genomic DNA region from the founder was amplified and sequenced using PCR primer pairs of: Tmprss3F, GGAGATCCCACATCTCTCACC; and Tmprss3R, AAATGCTATGCACCTACATCAAC. After the mutations were confirmed by sequencing, founder mice carrying the mutations were mated to generate wild-type (WT), heterozygous, and homozygous mutation mice. A representative mouse genotyping determined by Sanger sequencing is shown in Figure 1C.

Tmprss3A306T/A306T mice display late-onset progressive HL

With the aim of rescuing auditory function in the Tmprss3A306T/A306T mice, we first performed auditory brain stem response (ABR) (representing sound-evoked neural output of the cochlea) and distortion product otoacoustic emission (DPOAE) (measurement of outer hair cell function) tests to study hearing in Tmprss3A306T/A306T mice (Figures 1D–1G and S1). We found that there was no change in ABR/DPOAE thresholds or ABR wave1 amplitudes in the Tmprss3A306T/A306T ears compared with the WT CBA/CaJ ears across all frequencies at 1.5 and 10.5 months (Figures 1D, 1E, and S1A–S1D). By 16.5 months, the ABR thresholds were significantly elevated in the Tmprss3A306T/A306T ears by an average of 15 dB compared with WT ears across most frequencies (Figure 1F). DPOAE thresholds were significantly elevated at 45 kHz and wave 1 amplitudes showed significant reductions at 32 kHz with 80- and 90-dB sound pressure level (SPL) stimulation (Figures S1E and S1F). At 22.5 months, ABR thresholds were significantly elevated by an average of 26 dB across all frequencies (Figure 1G). We did not find changes in wave 1 amplitudes in Tmprss3A306T/A306T ears compared with WT ears at 22.5 months (Figure S1H), as both were much lower than that at 16.5 months (Figure S1F), an indication that neuronal activities were reduced in the WT mice due to aging. The DPOAE thresholds were elevated in Tmprss3A306T/A306T ears compared with WT ears at 22.5 months (Figure S1G). These results showed late-onset progressive HL that started after 10.5 months with significantly elevated ABR/DPOAE thresholds and reduction in wave 1 amplitudes at 16.5 months, which further deteriorated by age 22.5 months.

Designing an AAV-TMPRSS3 gene therapy strategy

The phenotype of late onset and progressive HL in the Tmprss3A306T/A306T KI mice supports the development of a gene therapy strategy to rescue hearing in adult mouse cochleae that is clinically relevant. We chose to use AAV2 for delivery to evaluate gene therapy in the Tmprss3A306T/A306T mice as AAV2 transduces the inner hair cells (IHCs) and the outer hair cells (OHCs) with high efficiency: AAV2 transduced all the IHCs and a majority of OHCs in the apex and mid-turn with a slight reduction in the base turn in adult mouse cochlea (Figure S2). The size of the TMPRSS3 gene is less than 1.5 kb, making it ideal for AAV-mediated delivery strategy. We cloned the coding sequences (CDSs) of mouse-Tmprss3 (mTmprss3) into the AAV2 backbone to produce AAV2-mTmprss3 (Figure 2A). To evaluate the potential functional ototoxicity of AAV2-mTmprss3, we injected the AAV2-mTmprss3 into the 1-month-old WT mice and tested hearing by ABR and DPOAE 4 weeks later. To our surprise, we found that ABR thresholds were significantly elevated by an average of 35 dB across all frequencies in AAV2-mTmprss3-injected ears compared with uninjected ears (Figure 2B). Similarly, there was a complete loss of DPOAE across frequencies in the injected ears (Figure 2C). Significant ABR threshold shifts and the loss of DPOAE indicated major damage by AAV2-mTmprss3 to the mouse inner ear. To study the cellular ototoxicity by AAV2-mTmprss3, we quantified the number of OHCs and IHCs by whole-mount labeling of AAV2-mTmprss3-injected ears and uninjected contralateral ears. There was a slight reduction in IHCs in AAV2-mTmprss3-injected ears compared with uninjected ears, especially in the middle and basal turns (Figures 2D and 2F). There was a significant reduction in the number of OHCs across the entire cochlear turn, with the base turn being most severely affected in AAV2-mTmprss3-injected ears (Figures 2E and 2G). These results demonstrate that AAV2-mTmprss3 induces ototoxicity by preferentially damaging OHCs, resulting in profound HL.Figure 2 AAV2-hTMPRSS3 gene therapy design strategy

(A) The construct of AAV-mTmprss3. (B) ABR thresholds were significantly elevated in WT mice ears injected with AAV2-mTmprss3 (red) compared with uninjected ears (blue). Time of injection: age 1 month of age. Time of testing: age 2 months. (C) DPOAE thresholds were no longer detectable in WT mice ears injected with AAV2-mTmprss3 (red) compared with uninjected ears (blue). Time of injection: age 1 month. Time of testing: age 2 months. (D) In an injected WT cochlea, major OHC loss and some IHC loss were seen in the middle-base turn, whereas some OHC loss was seen in the apical-middle turn. MYO7A labels HC. (E) In the uninjected contralateral ear, a full set of OHCs and IHCs were seen along the cochlear turns. Scale bar, 50 μm. (F and G) Quantification of IHCs (F) and OHCs (G) per 400-μm section for three injected (red) and uninjected (blue) cochleae from three mice. Time of injection: age 1 month. Time of imaging: age 2 months. (H) The construct of AAV-hTMPRSS3. (I) ABR thresholds in WT mice ears injected with AAV2-hTMPRSS3 (red) compared with uninjected ears (blue). Time of injection: age 2 months. Time of testing: age 3 months. (J) DPOAE thresholds in WT mice ears injected with AAV2-hTMPRSS3 (red) compared with uninjected ears (blue). Time of injection: age 2 months. Time of testing: age 3 months. (K) A full set of OHCs and IHCs were seen in an AAV2-hTMPRSS3-injected WT ear along the cochlear turn. (L) The contralateral uninjected inner ear showed a full set of OHCs and IHCs. Scale bar, 50 μm. (M and N) Quantification of IHCs (M) and OHCs (N) per 400-μm section for three injected (red) and uninjected (blue) cochleae from three mice. Time of injection: age 2 months. Time of imaging: age 3 months. Values and error bars reflect mean ± SEM.

While we do not know the origin of the toxicity mediated by mouse Tmprss3, one approach to circumvent the issue is to test the use of the human TMPRSS3 (hTMPRSS3) gene for delivery, as the human TMPRSS3 gene will be used in the clinic. We constructed an AAV2-hTMPRSS3 (Figure 2H) and injected it into 2-month-old WT mouse inner ears and performed hearing tests and inner ear characterization 4 weeks later. In the injected ears, the ABR threshold showed a similar profile to uninjected inner ears (Figure 2I). DPOAE thresholds in the injected ears were slightly elevated but were not significantly greater than uninjected ears (Figure 2J). There was no change in the number of IHCs and OHCs in AAV2-hTMPRSS3-injected compared to uninjected ears (Figures 2K–2N). Taken together, we conclude that the AAV2-hTMPRSS3 inner ear delivery did not impair normal hearing or damage HC, and thus AAV2-hTMPRSS3 is suitable for gene therapy in the Tmprss3A306T/A306T mouse model.

AAV2-mediated TMPRSS3 expression in HC and SGN of mouse cochlea

The expression of Tmprss3 gene or the localization of TMPRSS3 protein is not well studied in adult mouse inner ear. To study the expression pattern of Tmprss3 in adult mouse inner ear and to evaluate the expression of the human TMPRSS3 gene as the result of AAV2-mediated delivery, we performed RNA-FISH (fluorescence in situ hybridization) (RNAscope)50,51 to detect mouse Tmprss3 and human TMPRSS3 genes, respectively. In WT mouse inner ear, a probe against the mouse Tmprss3 mRNA showed strong hybridization to the endogenous Tmprss3 mRNA that was primarily in the cochlear HC (MYO7A+) in the sensory epithelium (Figure 3A). We then used a human TMPRSS3 probe for RNAscope in WT mouse inner ear and did not detect significant signals above the background (Figure 3B). We quantified the hybridization signals by ImageJ. In WT mouse inner ear, the endogenous Tmprss3 mRNA was expressed at a higher level in the IHCs than the OHCs, whereas no human TMPRSS3 signals were detected (Figure 3E). Our data show that the human TMPRSS3 probe does not cross-react with the mouse Tmprss3 mRNA, demonstrating the specificity of the human TMPRSS3 probe.Figure 3 AAV2-hTMPRSS3 inner ear delivery recapitulated endogenous Tmprss3 expression in HC in Tmprss3A306T/A306T KI mice

(A) Endogenous mouse Tmprss3 gene expression in WT mouse cochlea was detected by a mouse Tmprss3 probe using RNAscope in a pattern overlapping with the inner and outer hair cells (IHCs and OHCs). (B) In WT mouse cochlea, a human TMPRSS3 probe showed very low or no cross-reactivity with the endogenous mouse Tmprss3 mRNA. (C) In a Tmprss3A306T/A306T mouse cochlea injected with AAV2-hTMPRSS3, a human TMPRSS3 probe detected the transgene expression of the human TMPRSS3 mRNA in the IHCs and OHCs. (D) In a Tmprss3A306T/A306T mouse cochlea without injection, a human TMPRSS3 probe did not detect mRNA above the background by RNAscope. Positive signals by RNAscope are shown as red punctuates on a black background. HC were labeled with MYO7A. The white square marks the area with higher magnification for better visualization. Scale bars, 10 μm. (E) Quantification of mouse Tmprss3 or human TMPRSS3 mRNA per IHC and OHC by RNAscope. Data are from five z stack images obtained from the apex and/or apex-mid turn regions of each sample group (control n = 3, AAV n = 2). Results are expressed as an average number of dots per IHC and OHC and are displayed as a scatter column plot with medians indicated by the horizontal bar. p values are shown on top of each dashed line between compared groups. Time of injection: age 80 weeks. Time of imaging: age 86 weeks. The RNAscope images were optimized individually for display purposes. Unpaired t-Test with Welch’s correction was used to compare average number of puncta per cell between samples. Data represent mean ± SEM. Significance threshold was set as p < 0.05.

We then injected AAV2-hTMPRSS3 into the Tmprss3A306T/A306T mouse inner ear and performed RNAscope. RNAscope showed that the TMPRSS3 mRNA signals became clearly visible in the HC (Figure 3C) compared with uninjected cochlea, where no signal was detected (Figure 3D) for the same animal by the same probe. Quantification showed that a similar level of human TMPRSS3 transgene expression was detected in the IHCs and OHCs of injected Tmprss3A306T/A306T (Figure 3E), whereas the overall TMPRSS3 transgene expression level was lower than the endogenous Tmprss3 mRNA in WT mouse inner ear (Figures 3A, 3C, and 3D). This study confirmed species-specific probes for TMPRSS3/Tmprss3 mRNA analyses by the RNAscope method.

The failure of cochlear implant in some TMPRSS3 patients suggests that the SGN were affected by TMPRSS3 mutations. However, it is not known if TMPRSS3 is expressed in the adult SGN or if any defect in the SGN is the direct result of the lack of TMPRSS3 expression in the neurons or an indirect result of other cells that lack TMPRSS3 expression. We examined the RNAscope data in the modiolus region that houses the SGN . RNAscope showed a distinct expression pattern of endogenous Tmprss3 mRNA by the mouse Tmprss3 probe in the modiolus region with the level lower than in HC (Figure 4A). Combined with the immunofluorescence staining of HuD proteins to visualize SGN soma, the Tmprss3 expression largely overlapped with the SGN (Figure 4A). Signals were also detected in the region outside the SGN , suggesting that Tmprss3 is expressed in other cell types in the modiolus region such as Schwann cells and satellite cells (Figure 4A). Again, the human TMPRSS3 probe did not show signals above the background in the WT mouse modiolus region (Figure 4B), consistent with the specificity of the probe against human TMPRSS3 mRNA.Figure 4 AAV2-hTMPRSS3 inner ear delivery recapitulated endogenous Tmprss3 expression in SGN in Tmprss3A306T/A306T KI mice

(A) In the WT mouse modiolus, endogenous mouse Tmprss3 gene expression was detected by a mouse Tmprss3 probe using RNAscope in a pattern largely overlapping with the SGN (HuD, dotted cycle). (B) In the WT mouse modiolus, a human TMPRSS3 probe showed no cross-reactivity with the endogenous mouse Tmprss3 mRNA. (C) In Tmprss3A306T/A306T mouse modiolus injected with AAV2-hTMPRSS3, a human TMPRSS3 probe detected the transgene expression of the human TMPRSS3 mRNA in the SGN (HuD, dotted cycle). (D) In Tmprss3A306T/A306T mouse modiolus without injection, a human TMPRSS3 probe did not detect mRNA above the background by RNAscope. Scale bars, 10 μm. (E) Quantification of mouse Tmprss3 or human TMPRSS3 mRNA in the modiolus region from the above-described samples. Data are from five z stack images obtained from the apex and/or apex-mid turn regions of each sample group (control n = 3, AAV n = 2). Results are expressed as an average number of dots per modiolus and are displayed as a scatter column plot with medians indicated by the horizontal bar. p values are shown on top of each dashed line between compared groups. Time of injection: age 80 weeks. Time of imaging: age 86 weeks. The RNAscope images were optimized individually for display purposes.Unpaired t-Test with Welch’s correction was used to compare average number of puncta per cell between samples. Data represent mean ± SEM. Significance threshold was set as p < 0.05.

We analyzed the RNAscope results of the human TMPRSS3 probe hybridized to the Tmprss3A306T/A306T inner ear injected with AAV2-hTMPRSS3. We detected the expression of the human TMPRSS3 transgene in the modiolus that overlapped with the SGN in AAV2-hTMPRSS3-injected Tmprss3A306T/A306T inner ear (Figure 4C), which coincided with the Tmprss3 expression pattern in WT mice, supporting that AAV2 targeted the SGN . In uninjected contralateral Tmprss3A306T/A306T inner ear, no TMPRSS3 expression was detected above the background (Figure 4D). Quantification confirmed that the human TMPRSS3 gene was expressed, although at a level lower than the endogenous Tmprss3 expression (Figure 4E). The RNAscope study showed that the endogenous Tmprss3 is expressed in the HC and SGN in WT inner ear, and AAV2-hTMPRSS3-mediated gene delivery targeted the same cell population in the Tmprss3A306T/A306T inner ears with the expression of the human TMPRSS3 gene.

AAV2-hTMPRSS3 gene therapy rescues hearing in Tmprss3A306T/A306T mice

We performed inner ear AAV2-hTMPRSS3 injection by round window injection with canal fenestration52 in Tmprss3A306T/A306T mice with an average age of 18.5 months, tested hearing 1 month after injection, and continued the testing monthly for 5 months (Figure 5A). Uninjected contralateral ears served as controls. WT CBA/CaJ mice at comparable ages were tested as additional controls to show the normal hearing profile in the mouse strain during aging. Representative ABR wave forms recorded from a Tmprss3A306T/A306T homozygous mutant mouse ear, an AAV2-hTMPRSS3-treated Tmprss3A306T/A306T ear, and a WT mouse ear were shown at age 20.5 months, using 11.32 kHz tone bursts at incrementally increasing SPLs from 20 to 90 dB (Figure S3A). One month after injection, overall lower ABR thresholds were detected in injected ears compared with uninjected control ears at frequencies of 5.66, 16, and 32 kHz (Figure 5B). For frequencies below 45.42 kHz, the ABR threshold reduction ranged from 6 dB at 11.32 kHz to 16 dB at 16 kHz, with an average reduction of 11 dB. No difference of DPOAE thresholds were detected between injected and uninjected ears (Figure S3B). The wave 1 amplitudes were statistically larger at 32 kHz above 50 dB SPL (Figure S3G). Two months after injection, ABR thresholds were further reduced at all frequencies in injected ears compared with uninjected control ears, ranging from 8 dB at 8 kHz to 18 dB at 45.24 kHz, with an average reduction of 15 dB across all frequencies (Figure 5C). Generally lower DPOAE thresholds were detected in injected compared with uninjected ears, with a significant reduction of 18 dB at 45.24 kHz (Figure S3C). At 3 and 4 months after injection, significant reductions in ABR thresholds were persistent in all frequencies in injected ears (Figures 5D and 5E), with an average reduction of 15 and 14 dB at 3 and 4 months, respectively. At the two time points, the DPOAE thresholds were generally lower and the wave 1 amplitudes at 32 kHz were larger in injected ears than in uninjected control ears (Figures S3D, S3E, 3I, and 3J). By 5 months post injection, with the exception of 16 kHz, ABR thresholds were still lower by an average of 9 dB (Figure 5F). Significant reduction in the DPOAE thresholds was detected at 8, 11.32, and 45.24 kHz (Figure S3F), and larger but not significant wave 1 amplitudes (Figure S3K) were seen in injected ears. At this age, the ABR thresholds in injected ears were indistinguishable from that of WT background CBA/CaJ strain at a comparable age of 23.5 months, which was elevated from that of CBA/CaJ of 22.5 months of age, an indication of ARHL in the CBA/CaJ background strain. In hearing tests of all ages, there was no significant difference of ABR thresholds between injected and WT CBA/CaJ ears (Figures 5B–5F), a demonstration of robust rescue of hearing by AAV2-hTMPRSS3 delivery in the Tmprss3A306T/A306T ears to a level similar to that in the WT control ears. Based on the data we conclude that a single administration of gene therapy by AAV2-hTMPRSS3 results in long-term hearing rescue in the Tmprss3A306T/A306T ears.Figure 5 AAV2-hTMPRSS3 injection in Tmprss3A306T/A306T mice rescues auditory function

(A) The schematic diagram of the experimental design. (B–F) ABR thresholds in Tmprss3A306T/A306T homozygous mice treated with AAV2-hTMPRSS3 ears (blue), untreated Tmprss3A306T/A306T homozygous ears (red), and WT CBA/CaJ mice (green) at 1 month after injection (B), 2 months after injection (C), 3 months after injection (D), 4 months after injection (E), and 5 months after injection (F), respectively. Significant hearing rescue was detected at all time points with the exception at 5 months post injection when WT CBA/CaJ mice started to exhibit hearing loss. Values and error bars reflect mean ± SEM.

AAV2-hTMPRSS3 promotes HC and SGN survival in Tmprss3A306T/A306T mice

Expression of Tmprss3 in HC and SGN suggests the requirement of TMPRSS3 in HC and SGN . To study the effect of the TMPRSS3 c.916G>A mutation on the inner ear and determine how AAV2-hTMPRSS3 gene delivery rescued the cellular phenotype, we performed immunolabeling of injected and uninjected cochleae and quantified IHCs, OHCs, and SGN . In uninjected Tmprss3A306T/A306T inner ear, a loss of OHCs in the basal and middle regions was detected (Figures 6A1–A3). In contrast, in injected Tmprss3A306T/A306T ear, more OHCs survived in the same regions (Figures 6B1–B3 and 6C). No significant difference was detected in IHC number between injected and uninjected control ear, although the IHC average was higher in the injected ear (Figures 6A1–B3 and 6D). In the base to mid-base turns of the modiolus region of uninjected Tmprss3A306T/A306T ear, the TuJ1 labeling was condensed and localized to one side of the SGN (Figures 6E1 and 6E2). In contrast, evenly distributed cytoplasmic labeling of TuJ1 was detected in the SGN of injected ears (Figures 6F1 and 6F2). Significantly, 3-fold more SGN were detected in the injected ear compared with the uninjected ear (Figure 6G). Taken together, these results demonstrate that the TMPRSS3 c.916G>A mutation caused the loss of the OHCs and SGN from base to mid-base at age 20.5 months and AAV2-hTMPRSS3 gene therapy rescued hair cell and SGN in the same regions.Figure 6 AAV2-hTMPRSS3 injection rescues HC and SGN in Tmprss3A306T/A306T cochleae

(A and B) Representative confocal images of cochleae from uninjected (A) and AAV2-hTMPRSS3-injected (B) ears of Tmprss3A306T/A306T mouse at age 82 weeks. The apex-middle (A1, B1), middle-base (A2, B2), and base (A3, B3) turns were dissected and stained with MYO7A (green) for HC. (C and D) Quantification and comparison of OHCs (C) and IHCs (D) in three representative regions of apex-middle, middle-base, and base of injected (blue) and uninjected (red) Tmprss3A306T/A306T cochleae. Significantly more OHC survived in injected ear compared with uninjected ear in the base and middle-base turns. (E and F) Representative confocal images of a modiolus cross-section through the whole cochlea from uninjected (E) and AAV2-hTMPRSS3-injected (F) ear of Tmprss3A306T/A306T mouse at age 82 weeks. The middle-base modiolus (E1, F1) and base modiolus (E2, F2) were stained with TuJ1 (red) for the SGN . Arrows point to enlarged insets with higher magnifications of the square. Dotted circles showed an example of a SGN soma with TuJ1 labeling, with condensed TuJ1 labeling (arrowhead) localized to one side within an SGN in an uninjected ear (arrow). (G) Quantification and comparison of TuJ1-positive SGNs between uninjected (red) and AAV2-hTMPRSS3-injected (blue) Tmprss3A306T/A306T cochleae. Values and error bars reflect mean ± SEM. n = 3. Scale bars, 25 μm.

Discussion

With the goal to develop gene therapy for DFNB8 in humans, we created a KI mouse model with a frequent human DFNB8 TMPRSS3 mutation that exhibits late onset and progressive HL similar to DFNB5 patients. We show that one-time local administration of AAV2-hTMPRSS3 gene therapy in aged KI mice of 18.5 months restores hearing at around age 2 years. Gene replacement therapy with the human TMPRSS3 promotes the survival of HC and SGNs, both of which are required for hearing and the latter is essential to the treatment outcome of cochlear implants.

TMPRSS3 is a type II membrane serine protease and is associated with cancer development.53,54 Tmprss3 has been found to be expressed in the developing cochlea including HC and SGN .55 The lack of TMPRSS3 has resulted in hair cell death in mouse organoids and SGN degeneration in vitro.56,57 Patients with homozygous TMPRSS3 mutations manifest either postlingual progressive HL (DFNB8) or congenital profound HL (DFNB10) (https://hereditaryhearingloss.org/recessive-genes),39,58 likely due to the nature of the mutations. In TMPRSS3 patients, cochlear implant is a standard form of treatment. Recently it has been suggested that, in some TMPRSS3 patients, the treatment outcome may diminish over time, leaving the patients without any treatment option.59,60 We aim to develop a gene therapy strategy to rescue hearing using a TMPRSS3 mouse model with the potential to be further developed for the clinic.

AAV is one of the most effective gene delivery vectors that transduce both dividing and non-dividing cells to provide long-term gene expression.61 In the inner ear, previous studies found that conventional AAVs could efficiently transduce IHCs but were less efficient in transducing OHCs.31,32 We previously reported that AAV2 harboring GFP transduces almost all the IHCs and a large number of OHCs in adult C57BL/6 mouse cochleae by canalostomy.62 This delivery route was also used in CBA/CaJ mouse cochleae to show efficient IHC and OHC transduction.63 In a recent study, it was demonstrated that AAV2 transgene expression is in both OHCs (84%) and IHCs (97%) in adult C3H/FeJ mouse cochleae using a round window membrane plus canal fenestration delivery method.64 Combined, these studies have indicated that animal strain, animal age, viral titer, viral promoter, and even viral processing and purifying procedure can affect transduction efficiency.52,65 In this study, the human TMPRSS3 gene was delivered into a mouse model by AAV2 in CBA/CaJ background using a round window membrane with canal fenestration delivery route, which showed robust transduction of IHCs and efficient transduction of OHCs.

Gene therapy has been extensively used to treat mouse models for human genetic HL with success.13,19,25,26,33,66 A majority of inner ear gene therapies, however, were only successful when performed in neonatal or young inner ears, not in adults.13,23 This could be due to the fact that many types of genetic HL are congenital and the inner ear cell types to be targeted are severely damaged or no longer available when mature, thus limiting the efficacy by intervention in the mature inner ear. Furthermore, the delivery vehicles may show differences targeting neonatal or adult cochlear cell types.10,62,67 In humans, even newborn inner ears are fully mature, so any gene therapy strategy for patients would require establishment of a window of opportunity for treatment and demonstration that successful hearing rescue can be achieved in fully mature animal inner ears. The Tmprss3 knockout mouse exhibits severe congenital hair cell loss and profound HL,39 and therefore is not a suitable model to develop TMPRSS3 gene therapy to rescue hearing in the mature inner ear. We chose to create a mouse model with a frequent founder human TMPRSS3 mutation c.916G>A (p.A306T) that causes postlingual progressive HL of DFNB8 in patients.42,44 To circumvent the mouse inbred strains such as C57 with accelerated ARHL, we chose to use CBA/CaJ as the background strain, so that the strain-dependent HL is not manifested until an advanced age, which allowed us to study late onset progressive HL and the treatment outcome in aged mice. The combination of mutation and mouse strain selections are important factors that enabled us to establish a KI mouse model of DFNB8 that is progressive HL. The model is ideal to test our gene therapy strategy to rescue hearing in not only mature but aged inner ear by one-time AAV administration.

It is surprising that the delivery of the mouse WT Tmprss3 copy into normal adult mouse HC leads to severe hair cell loss and profound HL. While the mechanism is not understood, it is likely that the level of the mouse TMPRSS3 protein is important to the survival of HC and essential for normal hearing. The detrimental effect of the mouse TMPRSS3 on HC and hearing prompted us to test the human TMPRSS3 gene. Again, it is to our surprise that the human TMPRSS3, despite sharing a high degree of homology with the mouse TMPRSS3, 88% identical on the protein sequence (Figure S4),68 does not cause hair cell loss or HL. Given our goal of using the human TMPRSS3 gene for the clinical development, we have thus chosen the human TMPRSS3 for gene therapy intervention in the KI Tmprss3 mouse model.

Due to the lack of a suitable anti-TMPRSS3 antibody, we carried out RNA-FISH (RNAscope) to study endogenous Tmprss3 expression in mouse inner ear and evaluate the inner ear cells targeted by AAV2-hTMPRSS3. In the mouse, Tmprss3 has been shown to be expressed in HC and other cochlear cell types including the cells in the modiolus region during development.39,58 A recent study also showed low TMPRSS3 expression in human auditory neurons.58

By RNAscope, we found that AAV2 carrying the human TMPRSS3 CDS can properly target IHCs, OHCs, and SGNs. The TMPRSS3 transgene expression pattern resembles that of the WT Tmprss3 expression. The data strongly support our approach of AAV2 delivery in the mouse inner ear to target the cells that are directly affected by the lack of Tmprss3 expression and evaluate the rescue effect. A few studies have reported that the outcomes of CI patients with TMPRSS3-related HL were inconsistent.40,43,59,60,69,70 Our results could provide a promising therapeutic avenue for those patients who have poor CI outcomes.

The mouse inner ear still undergoes development until the onset of hearing at around P12.71 However, in the clinic, it is impractical to perform genetic therapies at the corresponding ages in humans as the intervention would have to be carried out in utero. To translate these findings of preclinical mouse studies to humans, gene therapy research imperatively needs to focus on practical therapeutic strategies to perform in adult and aged mice. Among successful hearing rescue in animal models by gene or editing therapy, the vast majority of studies were carried out in the neonatal or infant stage.13,14,15,16,17,19,21,22,23,24,25,29,32,35,72,73 In Tmprss3A306T/A306T mice, HL does not start until age ∼16.5 months, thus the intervention in aged mouse inner ears would be ideal to determine the treatment effect. We started injection in the animals of an average age of 18.5 months and measured hearing 1 month after injection and continued the testing monthly for 5 months. Significant hearing rescue by the reduction in the ABR thresholds was detected 1 month after injection at some frequencies (Figure 5B). The rescue effect extended to all frequencies with increasing magnitudes (Figures 5C–5E). Overall, there is a decrease in average DPOAE thresholds and greater wave 1 amplitudes in injected compared with uninjected contralateral control Tmprss3A306T/A306T ears (Figure S3), supporting the improved OHC function and neuronal activities. The improved hearing was detected 4 months after injection at an age of 22.5 months. At 5 months post injection with average age of mice at 23.5 months, even the WT control mice of the same age started to exhibit ARHL, with an ABR threshold average that was indistinguishable from injected TMPRSS3A306T/A306Tmice (Figure 5F). We conclude that the strain-specific ARHL obscured the treatment effect by AAV2-hTMPRSS3 injection at 23.5 months. Our study demonstrates that one-time administration of AAV gene therapy in aged mice is sufficient to rescue hearing long term in mice due to a TMPRSS3 A306T mutation. The TMPRSS3 A306T is a common mutation that has been described in German, Dutch, Korean, and Chinese families.40,41,42,43,44 While our gene therapy has shown efficacy in treating the Tmprss3A306T/A306T mouse model, the same AAV2-hTMPRSS3 is applicable to all DFNB8 patients with TMPRSS3 mutations. Hearing rescue by local injection at an advanced age in mice offers a wide therapeutic window that could enable the intervention in patients with DFNB8.

In the Tmprss3A306T/A306T mice, significant hair cell, especially OHC, and neuronal loss was detected (Figure 6). In contrast to the Tmprss3 knockout mice in which HC die rapidly,39 our model with the A306T mutation showed a gradual hair cell and neuronal loss, which is consistent with late onset and progressive HL. After AAV2-hTMPRSS3 injection, both cell types survived significantly better, which is consistent with those cells targeted by AAV2. Interestingly, we also observed that TuJ1, a neuronal marker, showed an aggregation-like pattern with condensed TuJ1 signal restricted to neuron cell bodies unilaterally in the Tmprss3A306T/A306T mice, which may further support previous findings showing cellular and molecular basis of SGN degeneration and the subsequent decline of the auditory nerve function in presbyacusis.74 While the physiological outcomes of TuJ1 aggregation is unknown, both the altered TuJ1 localization and the neuronal loss strongly support the direct impact of the TMPRSS3 mutation A306T on the SGN and the rescue effect by AAV2-hTMPRSS3 local delivery. Mutations in TMPRSS3 cause DFNB8 with HL that is postlingual and progressive, and DFNB10 with congenital profound HL. For DFNB8 patients, it is highly likely that HC and SGNs are present after birth, and they degenerate overtime. For DFNB8 patients, early intervention by AAV2-hTMPRSS3 gene therapy may be used as a standalone therapy to rescue HC and SGNs to prevent HL. For TMPRSS3 patients with profound HL, their HC may have severely degenerated, and CI is the only option for treatment. For those patients, AAV2-hTMPRSS3 gene delivery can be used to promote SGN survival to enhance long-term CI treatment outcomes. In addition to HL caused by homozygous or compound heterozygous TMPRSS3 mutations, there has been an increase in reports that heterozygous TMPRSS3 mutations could contribute to accelerated ARHL75 in conjunction with a compound heterozygous mutation in another deafness gene.76,77 It is thus conceivable that gene therapy for TMPRSS3 developed in this study could be expanded to rescue hearing in a subset of patients exhibiting ARHL with the contribution of TMPRSS3 mutations. The study supports that a single administration of AAV gene therapy for TMPRSS3 in fully mature and aged inner ears could achieve noticeable restoration of hearing with long-term therapeutic effect. It is the proof-of-principle demonstration that gene therapy could be successfully implemented at a late stage in life and builds a solid foundation for its future clinical application.

Materials and methods

Animals

Tmprss3A306T/A306T mutant mice were generated at the Mouse Genome Engineering Core Facility of University of Nebraska Medical Center. The background was CBA/CaJ strain (Jackson Laboratory stock no. 000656) to eliminate the effect of ARHL. The mice were housed in groups of two to five per cage and allowed free access to food and water. The animals were maintained under standard conditions (room temperature [RT]: 22°C ± 2°C; relative humidity: 55% ± 10%) on a light:dark cycle of 12:12 h (6:00 a.m. to 6:00 p.m.). All procedures were approved by the Massachusetts Eye & Ear IACUC committee (protocol no. 08-04-008) and University of Miami Institutional Animal Care Committee (protocol no. 19–104), following the National Institutes of Health (NIH) Guidelines. All mouse experiments were performed in accordance with NIH guidelines for use and care of laboratory animals and were approved by the Massachusetts Eye & Ear IACUC committee (protocol no. 08-04-008).

Recombinant DNA constructs and AAV virus production

To construct recombinant AAV2 plasmids, CDSs for human TMPRSS3 (GenBank: accession no. BC074846; CDS: 1,362 bp, 453 aa) and mouse Tmprss3 (Genbank: accession no. NM_001163776; CDS: 1,428 bp, 475 aa) were individually inserted in the host vector linearized by NotI and BamHI enzymes. CDSs amplified by PCR were subcloned downstream of the hCMV enhancer/promoter in the host vector via In-Fusion HD enzyme (Takara), which fuses PCR-generated amplicons and linearized vector precisely by recognizing a 15 bp overlap at their ends. All constructs were sequenced prior to transfections to ensure no mutations were introduced during cloning. All the plasmids were propagated in DH5a E. coli cells and plasmids were extracted using endo-free plasmid purification kits (QIAGEN).

For recombinant AAV production, transgene vectors were packaged at the University of Massachusetts Medical School, Viral Vector Core, as described previously.78 In brief, HEK293 cells maintained in DMEM with GlutaMAX, penicillin/streptomycin, and 10% FBS were co-transfected with packaging plasmid (pRep2/Cap2; Agilent Technologies), adenovirus helper plasmid (pHelper; Agilent Technologies), and rAAV plasmid carrying a human or mouse full-length Tmprss3 expression cassette flanked by AAV2 ITRs. The pRep2/Cap2 expresses regulatory and capsid proteins of AAV2 serotype, which excises the recombinant genome from the rAAV vector plasmid, replicates the viral ssDNA genome, and packages the genome into AAV virions. Adenovirus E2A, E4, and viral-associated RNAs expressed from the pHelper provide helper functions essential for rAAV rescue, replication, and packaging.79 HEK293 cells were harvested 72 h post-transfection, suspended in a lysis buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 1 mM MgCl2, and lysed by three freeze-thaw cycles. Unencapsidated plasmid DNA was digested using 250 U Benzonase (Sigma-Aldrich), and virions were purified by ultracentrifugation on standard CsCl gradient and desalted by dialysis. The vectors are quality control tested by ddPCR titration for DNase-resistant vector genome concentration using probe and primers targeting the bGH region, and AAV purity was assessed by 4%–12% SDS-acrylamide gel electrophoresis followed by silver staining (Invitrogen). Viral genomic titers for AAV2.CMV.hTMPRSS3 and AAV2.CMV.mTmprss3 were calculated as 1.1 × 10E-13 and 1.0 × 10E-13 GC/mL, respectively.

Animal surgery

Tmprss3A306T/A306T and WT mice of either sex were anesthetized using intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg). The post-auricular incision was exposed by shaving and disinfected using 10% povidone iodine. The AAV2-mTmprss3 and AAV2-hTMPRSS3 were injected into the inner ears of WT and Tmprss3A306T/A306T mice. The AAV2-GFP was injected into the inner ears of WT mice. The total volume for each injection was 1 μL virus per cochlea.

Auditory brainstem response and DPOAE

Mice of either sex were anesthetized under the same conditions as for surgery. For ABR measurements, subcutaneous needle electrodes were inserted at the vertex (reference), ventral edge of the pinna (active electrode), and a ground reference near the tail. In a soundproof chamber, mice were presented with 5-ms tone pips (delivered at 35/s). The response was amplified 10,000-fold, then filtered (100 Hz–3 kHz band-pass), digitized, and averaged (1,024 responses) at each SPL. The sound level was elevated in 5-dB steps from 20 up to 90 dB SPL at stimuli of 5.66–45.24 kHz frequencies (in half-octave steps). The threshold and wave 1 amplitude were identified as described previously.17 During the same recording session, DPOAEs were measured under the same conditions as for ABRs. In brief, two primary tones (f2/f1 = 1.2) were set, with f2 varied between 5.66 and 45.24 kHz in half-octave steps. Primaries were swept from 20 to 80 dB SPL (for f2) in 5-dB steps. Thresholds required to produce a DPOAE at 5 dB SPL were computed by interpolation as f2 level. The information on the animals studied including the age of injection, the age of hearing tests, and the age of inner ear harvest is provided in Table S1.

Confocal microscopy and cell counting

Injected and uninjected cochleae were harvested from mice euthanized by CO2 inhalation. Temporal bones were fixed in 4% paraformaldehyde at 4°C overnight, then decalcified in 120 mM EDTA for at least 7 days until the tissues softened. After decalcification, the cochleae were dissected for whole-mount immunostaining or cryosection at 10 μm thickness using published methods.17,80,81 Tissues were infiltrated with 0.25% Triton X-100 and blocked with donkey serum (5%) for 1 h at RT, followed by washing 3 × 10 min with PBS and then incubated with primary antibody. Rabbit anti-MYO7A (1:500; Proteus BioSciences, no. 25-6790), mouse anti-beta III tubulin (1:250; BioLegend, no. 801201), and chicken anti-GFP (1:1000; Abcam, ab13970) were used overnight. Tissues were then washed for 3 × 10 min with PBS and the secondary antibodies were incubated for 1 h (donkey anti-rabbit IgG Alexa Fluor Plus 488, 1:1,000; donkey anti-mouse IgG Alexa Fluor Plus 594, 1:1,000; Thermo Fisher Scientific). Following secondary antibody incubation, tissues were washed for 3 × 10 min with PBS. Finally, tissues were placed on a microscope slide and mounted with VECTASHIELD antifade mounting medium containing DAPI (Vector Laboratories, no. H-1200). Images were taken with a Leica SP8 confocal laser scanning microscope (Leica Microsystems, Germany) via a 20× or 63× glycerin immersion lens. Images were edited by ImageJ software and tools in ImageJ were used for counting of HC and SGNs. For hair cell counting, Myosin7a-positive HC per 100 μm length were calculated in the apical, middle, and basal turns of cochleae. We counted at least two 100-μm segments from at least three independent cochleae. For SGN cell counting, TUJ1-positive cells were calculated in the SGN area. The average cell number per 104 μm2 was calculated for data analysis.

RNA-FISH, immunohistochemistry, and data quantification

RNA-FISH was performed using an RNAscope Multiplex Fluorescent Reagent Kit v.2 (ACD, no. 323110), and the protocol described here is adapted from Huang and Eckrich50,51 with minor modifications.50,51 In brief, membranous labyrinths were removed, and cochleae were immediately soaked in ice-cold 4% PFA in PBS. Cochleae were fixed for 2 h at RT on a shaker and washed for 3 × 10 min in 0.1% Tween 20 in PBS (PBT20), and subsequently dehydrated in a graded MeOH series (50%, 75%, and 100% in PBT20, 10 min for each grade). At the same time, probes were pre-warmed to 40°C for 10 min to allow aggregates to dissolve and then cooled down to RT. Protease III solution and Amplifiers I-IV were allowed to equilibrate to RT.

Before hybridization, cochleae were rehydrated at RT in a reverse MeOH series (100%, 75%, and 50% in PBT20; 10 min for each grade) and washed 6 × 5 min in PBT20. During the last washing step, the apical turn was dissected in PBT20 in a 35-mm sterile dish, and the stria vascularis and spiral ligament were carefully removed. Reissner’s membrane and the tectorial membrane were also separated from the sensory epithelium with fine forceps while the SGN was maintained intact. Dissected tissues collected from different sample groups were placed in individual mini cell strainers (pluriStrainer Mini 40 μm) containing 1 mL of PBT20, which were then transferred manually between the wells of a 24-well plate during all incubation and washing steps.

Dissected cochlea pieces were digested for 12 min at RT in 300 mL of Protease III solution. Following washing with 1 mL of PBT20 for 6 × 5 min to remove residual proteases, samples were incubated with hybridization probes for 2 h at 40°C. We used target probes against mouse TMPRSS3 mRNA (ACD, no. 553861) or human TMPRSS3 mRNA (ACD, no. 524691-C2). Subsequently, tissues were washed for 6 × 5 min with 1 mL of ACD washing buffer, re-fixed for 10 min with 4% PFA, and washed again for 6 × 5 min. Probe signals were amplified by incubation at 40°C in 4–5 drops of Amp1 (35 min), Amp2 (20 min), Amp3 (35 min), and Amp4 “Alt-A″” solution (20 min), respectively. Following each step, tissues were washed for 6 × 5 min in 1 mL of ACD washing buffer in mini cell strainers housed in a 24-well plate.

For a more precise identification of different cell types, the RNAscope assay was coupled to immunohistochemistry. In brief, tissues were permeabilized for 10 min with 0.5% Triton X-100 in PBS (PBST), and blocked for 1 h with 10% donkey serum in PBST at RT. Subsequently, samples were incubated with primary antibodies against MYO7A (polyclonal rabbit, 1:250, Proteus Biosciences, no. 25-6790) and HuD (monoclonal mouse, 1:100, SantaCruz, no. sc-48421) in antibody incubation buffer (5% donkey serum and 0.25% PBST) for 1 h at RT. Following washing with 0.1% PBST for 3 × 5 min, tissues were incubated at RT for 1 h with appropriate secondary antibodies (donkey α-rabbit IgG Alexa Fluor Plus 594, 1:500; donkey α-mouse IgG Alexa Fluor Plus 647, 1:500; Thermo Fisher Scientific) in the antibody incubation buffer. Following secondary antibody incubation, tissues were washed for 3 × 5 min with 0.1% PBST. Finally, tissues were placed on a microscope slide and mounted with Prolong Gold DAPI antifade medium (Invitrogen, no. P36931), and allowed to penetrate and solidify overnight before imaging.

RNAscope plus immunohistochemistry experiments were repeated in more than two animals per group. For each sample group, ≥5 z stack images were acquired using a 63× oil objective with 2× digital zoom in a Leica SP8 confocal laser scanning microscope. Fluorescently labeled mRNA was quantified using the particle analysis tool in ImageJ, which requires 8-bit grayscale images. After adjusting thresholds to remove background signal for each image, the average number of mRNA molecules per hair cell was determined by dividing the number of particles by the number of MYO7A-positive HC or by the total number of particles per HuD-positive SGN . For counting of IHCs, OHCs, and SGNs, we acquired z stacks by maximum intensity projections. The average number of IHCs and OHCs per hair cell was determined by dividing the number of particles by the number of MYO7A-positive HC or by the total number of particles per HuD-positive SGN .

Statistical analysis

We used GraphPad Prism (v.9, GraphPad Software, La Jolla, CA) for statistical analysis. For multiple comparisons in terms of ABRs and DPOAEs, statistical analyses were carried out by two-way ANOVA with Bonferroni corrections. For a numeric representation of RNAscope data, scatter column graphs were created using five z stack images. For comparisons between two groups, data were analyzed by two-tailed unpaired t tests with Welch’s correction. Significance threshold was set as p < 0.05 (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).

Data availability

All study data are included in the article and/or supplemental information.

Supplemental information

Document S1. Figures S1–S4

Table S1. The information on the animals studied including the age of injection, the age of hearing tests, and the age of inner ear harvest

Document S2. Article plus supplemental information

Acknowledgments

This work was supported by 10.13039/100000002 NIH R01DC016875, UG3TR002636, and UH3TR002636 (to Z.-Y.C.), R01DC019404 (to X.L. and Z.-Y.C.), Ines-Fredrick Yeatts Fund (to Z.-Y.C.), R01DC012115, R01DC005575, and 10.13039/100000005 DOD RH220053 (to X.L.). We acknowledge the BioRender that was used to create the graphical abstract. We acknowledge Olivia Catalini for help with the illustration.

Author contributions

H.S., X.L., and Z.-Y.C. supervised the project. W.D., V.E., H.S., X.L., and Z.-Y.C. designed the experiments. W.D., V.E., C.L., M.H., S.S., A.M.A., Z.H., C.B.G., and H.S. conducted the experiments. All authors analyzed data. W.D., V.E., C.L., and Z.-Y.C. wrote the manuscript. All authors reviewed and edited the manuscript.

Declaration of interests

Z.-Y.C. is a co-founder and a SAB member of Salubritas Therapeutics. X.L. and H.S. are scientific advisors to Rescue Hearing Inc.

Supplemental information can be found online at https://doi.org/10.1016/j.ymthe.2023.05.005.
==== Refs
References

1 World Health Organization Deafness and hearing loss https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss 2020
2 Lin F.R. Yaffe K. Xia J. Xue Q.L. Harris T.B. Purchase-Helzner E. Satterfield S. Ayonayon H.N. Ferrucci L. Simonsick E.M. Health ABC Study Group Hearing loss and cognitive decline in older adults JAMA Intern. Med. 173 2013 293 299 23337978
3 Thomson R.S. Auduong P. Miller A.T. Gurgel R.K. Hearing loss as a risk factor for dementia: a systematic review Laryngoscope Investig. Otolaryngol. 2 2017 69 79
4 Géléoc G.S.G. Holt J.R. Sound strategies for hearing restoration Science 344 2014 1241062
5 Müller U. Barr-Gillespie P.G. New treatment options for hearing loss Nat. Rev. Drug Discov. 14 2015 346 365 25792261
6 Lenz D.R. Avraham K.B. Hereditary hearing loss: from human mutation to mechanism Hear. Res. 281 2011 3 10 21664957
7 Shearer A.E. Hildebrand M.S. Sloan C.M. Smith R.J.H. Deafness in the genomics era Hear. Res. 282 2011 1 9 22016077
8 Lin X. Tang W. Ahmad S. Lu J. Colby C.C. Zhu J. Yu Q. Applications of targeted gene capture and next-generation sequencing technologies in studies of human deafness and other genetic disabilities Hear. Res. 288 2012 67 76 22269275
9 Pillay S. Zou W. Cheng F. Puschnik A.S. Meyer N.L. Ganaie S.S. Deng X. Wosen J.E. Davulcu O. Yan Z. Adeno-associated virus (AAV) serotypes have distinctive interactions with domains of the cellular AAV receptor J. Virol. 91 2017 e00391-17 28679762
10 Landegger L.D. Pan B. Askew C. Wassmer S.J. Gluck S.D. Galvin A. Taylor R. Forge A. Stankovic K.M. Holt J.R. Vandenberghe L.H. A synthetic AAV vector enables safe and efficient gene transfer to the mammalian inner ear Nat. Biotechnol. 35 2017 280 284 28165475
11 Russell S. Bennett J. Wellman J.A. Chung D.C. Yu Z.F. Tillman A. Wittes J. Pappas J. Elci O. McCague S. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial Lancet 390 2017 849 860 28712537
12 Day J.W. Finkel R.S. Chiriboga C.A. Connolly A.M. Crawford T.O. Darras B.T. Iannaccone S.T. Kuntz N.L. Peña L.D.M. Shieh P.B. Onasemnogene abeparvovec gene therapy for symptomatic infantile-onset spinal muscular atrophy in patients with two copies of SMN2 (STR1VE): an open-label, single-arm, multicentre, phase 3 trial Lancet Neurol. 20 2021 284 293 33743238
13 Nist-Lund C.A. Pan B. Patterson A. Asai Y. Chen T. Zhou W. Zhu H. Romero S. Resnik J. Polley D.B. Improved TMC1 gene therapy restores hearing and balance in mice with genetic inner ear disorders Nat. Commun. 10 2019 236 30670701
14 Askew C. Rochat C. Pan B. Asai Y. Ahmed H. Child E. Schneider B.L. Aebischer P. Holt J.R. Tmc gene therapy restores auditory function in deaf mice Sci. Transl. Med. 7 2015 295ra108
15 Yoshimura H. Shibata S.B. Ranum P.T. Moteki H. Smith R.J.H. Targeted allele suppression prevents progressive hearing loss in the mature murine model of human TMC1 deafness Mol. Ther. 27 2019 681 690 30686588
16 Shibata S.B. Ranum P.T. Moteki H. Pan B. Goodwin A.T. Goodman S.S. Abbas P.J. Holt J.R. Smith R.J.H. RNA interference prevents autosomal-dominant hearing loss Am. J. Hum. Genet. 98 2016 1101 1113 27236922
17 Gao X. Tao Y. Lamas V. Huang M. Yeh W.H. Pan B. Hu Y.J. Hu J.H. Thompson D.B. Shu Y. Treatment of autosomal dominant hearing loss by in vivo delivery of genome editing agents Nature 553 2018 217 221 29258297
18 György B. Meijer E.J. Ivanchenko M.V. Tenneson K. Emond F. Hanlon K.S. Indzhykulian A.A. Volak A. Karavitaki K.D. Tamvakologos P.I. Gene transfer with AAV9-PHP.B rescues hearing in a mouse model of Usher syndrome 3A and transduces hair cells in a nonhuman primate Mol. Ther. Methods Clin. Dev. 13 2019 1 13 30581889
19 Dulon D. Papal S. Patni P. Cortese M. Vincent P.F. Tertrais M. Emptoz A. Tlili A. Bouleau Y. Michel V. Clarin-1 gene transfer rescues auditory synaptopathy in model of Usher syndrome J. Clin. Invest. 128 2018 3382 3401 29985171
20 Geng R. Omar A. Gopal S.R. Chen D.H.C. Stepanyan R. Basch M.L. Dinculescu A. Furness D.N. Saperstein D. Hauswirth W. Modeling and preventing progressive hearing loss in Usher syndrome III Sci. Rep. 7 2017 13480
21 Pan B. Askew C. Galvin A. Heman-Ackah S. Asai Y. Indzhykulian A.A. Jodelka F.M. Hastings M.L. Lentz J.J. Vandenberghe L.H. Gene therapy restores auditory and vestibular function in a mouse model of Usher syndrome type 1c Nat. Biotechnol. 35 2017 264 272 28165476
22 Lentz J.J. Jodelka F.M. Hinrich A.J. McCaffrey K.E. Farris H.E. Spalitta M.J. Bazan N.G. Duelli D.M. Rigo F. Hastings M.L. Rescue of hearing and vestibular function by antisense oligonucleotides in a mouse model of human deafness Nat. Med. 19 2013 345 350 23380860
23 Chien W.W. Isgrig K. Roy S. Belyantseva I.A. Drummond M.C. May L.A. Fitzgerald T.S. Friedman T.B. Cunningham L.L. Gene therapy restores hair cell stereocilia morphology in inner ears of deaf whirler mice Mol. Ther. 24 2016 17 25 26307667
24 Isgrig K. Shteamer J.W. Belyantseva I.A. Drummond M.C. Fitzgerald T.S. Vijayakumar S. Jones S.M. Griffith A.J. Friedman T.B. Cunningham L.L. Chien W.W. Gene therapy restores balance and auditory functions in a mouse model of Usher syndrome Mol. Ther. 25 2017 780 791 28254438
25 Akil O. Dyka F. Calvet C. Emptoz A. Lahlou G. Nouaille S. Boutet de Monvel J. Hardelin J.P. Hauswirth W.W. Avan P. Dual AAV-mediated gene therapy restores hearing in a DFNB9 mouse model Proc. Natl. Acad. Sci. USA 116 2019 4496 4501 30782832
26 Al-Moyed H. Cepeda A.P. Jung S. Moser T. Kügler S. Reisinger E. A dual-AAV approach restores fast exocytosis and partially rescues auditory function in deaf otoferlin knock-out mice EMBO Mol. Med. 11 2019 e9396
27 György B. Sage C. Indzhykulian A.A. Scheffer D.I. Brisson A.R. Tan S. Wu X. Volak A. Mu D. Tamvakologos P.I. Rescue of hearing by gene delivery to inner ear hair cells using exosome-associated AAV Mol. Ther. 25 2017 379 391 28082074
28 Kim M.A. Cho H.J. Bae S.H. Lee B. Oh S.K. Kwon T.J. Ryoo Z.Y. Kim H.Y. Cho J.H. Kim U.K. Lee K.Y. Methionine sulfoxide reductase B3-targeted in utero gene therapy rescues hearing function in a mouse model of congenital sensorineural hearing loss Antioxid. Redox Signal. 24 2016 590 602 26649646
29 Chang Q. Wang J. Li Q. Kim Y. Zhou B. Wang Y. Li H. Lin X. Virally mediated Kcnq1 gene replacement therapy in the immature scala media restores hearing in a mouse model of human Jervell and Lange-Nielsen deafness syndrome EMBO Mol. Med. 7 2015 1077 1086 26084842
30 Iizuka T. Kamiya K. Gotoh S. Sugitani Y. Suzuki M. Noda T. Minowa O. Ikeda K. Perinatal Gjb2 gene transfer rescues hearing in a mouse model of hereditary deafness Hum. Mol. Genet. 24 2015 3651 3661 25801282
31 Yu Q. Wang Y. Chang Q. Wang J. Gong S. Li H. Lin X. Virally expressed connexin26 restores gap junction function in the cochlea of conditional Gjb2 knockout mice Gene Ther. 21 2014 71 80 24225640
32 Akil O. Seal R.P. Burke K. Wang C. Alemi A. During M. Edwards R.H. Lustig L.R. Restoration of hearing in the VGLUT3 knockout mouse using virally mediated gene therapy Neuron 75 2012 283 293 22841313
33 Taiber S. Cohen R. Yizhar-Barnea O. Sprinzak D. Holt J.R. Avraham K.B. Neonatal AAV gene therapy rescues hearing in a mouse model of SYNE4 deafness EMBO Mol. Med. 13 2021 e13259
34 Delmaghani S. Defourny J. Aghaie A. Beurg M. Dulon D. Thelen N. Perfettini I. Zelles T. Aller M. Meyer A. Hypervulnerability to sound exposure through impaired adaptive proliferation of peroxisomes Cell 163 2015 894 906 26544938
35 Emptoz A. Michel V. Lelli A. Akil O. Boutet de Monvel J. Lahlou G. Meyer A. Dupont T. Nouaille S. Ey E. Local gene therapy durably restores vestibular function in a mouse model of Usher syndrome type 1G Proc. Natl. Acad. Sci. USA 114 2017 9695 9700 28835534
36 Bademci G. Foster J. Mahdieh N. Bonyadi M. Duman D. Cengiz F.B. Menendez I. Diaz-Horta O. Shirkavand A. Zeinali S. Comprehensive analysis via exome sequencing uncovers genetic etiology in autosomal recessive nonsyndromic deafness in a large multiethnic cohort Genet. Med. 18 2016 364 371 26226137
37 Sloan-Heggen C.M. Bierer A.O. Shearer A.E. Kolbe D.L. Nishimura C.J. Frees K.L. Ephraim S.S. Shibata S.B. Booth K.T. Campbell C.A. Comprehensive genetic testing in the clinical evaluation of 1119 patients with hearing loss Hum. Genet. 135 2016 441 450 26969326
38 Scott H.S. Kudoh J. Wattenhofer M. Shibuya K. Berry A. Chrast R. Guipponi M. Wang J. Kawasaki K. Asakawa S. Insertion of β-satellite repeats identifies a transmembrane protease causing both congenital and childhood onset autosomal recessive deafness Nat. Genet. 27 2001 59 63 11137999
39 Fasquelle L. Scott H.S. Lenoir M. Wang J. Rebillard G. Gaboyard S. Venteo S. François F. Mausset-Bonnefont A.L. Antonarakis S.E. Tmprss3, a transmembrane serine protease deficient in human DFNB8/10 deafness, is critical for cochlear hair cell survival at the onset of hearing J. Biol. Chem. 286 2011 17383 17397 21454591
40 Weegerink N.J.D. Schraders M. Oostrik J. Huygen P.L.M. Strom T.M. Granneman S. Pennings R.J.E. Venselaar H. Hoefsloot L.H. Elting M. Genotype-phenotype correlation in DFNB8/10 families with TMPRSS3 mutations J. Assoc. Res. Otolaryngol. 12 2011 753 766 21786053
41 Lee J. Baek J.I. Choi J.Y. Kim U.K. Lee S.H. Lee K.Y. Genetic analysis of TMPRSS3 gene in the Korean population with autosomal recessive nonsyndromic hearing loss Gene 532 2013 276 280 23958653
42 Chung J. Park S.M. Chang S.O. Chung T. Lee K.Y. Kim A.R. Park J.H. Kim V. Park W.Y. Oh S.H. A novel mutation of TMPRSS3 related to milder auditory phenotype in Korean postlingual deafness: a possible future implication for a personalized auditory rehabilitation J. Mol. Med. 92 2014 651 663 24526180
43 Elbracht M. Senderek J. Eggermann T. Thürmer C. Park J. Westhofen M. Zerres K. Autosomal recessive postlingual hearing loss (DFNB8): compound heterozygosity for two novel TMPRSS3 mutations in German siblings J. Med. Genet. 44 2007 e81 17551081
44 Gao X. Yuan Y.Y. Wang G.J. Xu J.C. Su Y. Lin X. Dai P. Novel mutations and mutation combinations of TMPRSS3 cause various phenotypes in one Chinese family with autosomal recessive hearing impairment Biomed. Res. Int. 2017 2017 4707315 28246597
45 Erway L.C. Shiau Y.W. Davis R.R. Krieg E.F. Genetics of age-related hearing loss in mice. III. Susceptibility of inbred and F1 hybrid strains to noise-induced hearing loss Hear. Res. 93 1996 181 187 8735078
46 Chen X. Abad C. Chen Z.Y. Young J.I. Gurumurthy C.B. Walz K. Liu X.Z. Generation and characterization of a P2rx2 V60L mouse model for DFNA41 Hum. Mol. Genet. 30 2021 985 995 33791800
47 Quadros R.M. Miura H. Harms D.W. Akatsuka H. Sato T. Aida T. Redder R. Richardson G.P. Inagaki Y. Sakai D. Easi-CRISPR: a robust method for one-step generation of mice carrying conditional and insertion alleles using long ssDNA donors and CRISPR ribonucleoproteins Genome Biol. 18 2017 92 28511701
48 Miura H. Quadros R.M. Gurumurthy C.B. Ohtsuka M. Easi-CRISPR for creating knock-in and conditional knockout mouse models using long ssDNA donors Nat. Protoc. 13 2018 195 215 29266098
49 Harms D.W. Quadros R.M. Seruggia D. Ohtsuka M. Takahashi G. Montoliu L. Gurumurthy C.B. Mouse genome editing using the CRISPR/Cas system Curr. Protoc. Hum. Genet. 83 2014 1 27
50 Wang F. Flanagan J. Su N. Wang L.C. Bui S. Nielson A. Wu X. Vo H.T. Ma X.J. Luo Y. RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues J. Mol. Diagn. 14 2012 22 29 22166544
51 Huang G. Eckrich S. Quantitative fluorescent in situ hybridization reveals differential transcription profile sharpening of endocytic proteins in cochlear hair cells upon maturation Front. Cell. Neurosci. 15 2021 643517
52 Yoshimura H. Shibata S.B. Ranum P.T. Smith R.J.H. Enhanced viral-mediated cochlear gene delivery in adult mice by combining canal fenestration with round window membrane inoculation Sci. Rep. 8 2018 2980 29445157
53 Li S.L. Chen X. Wu T. Zhang X.W. Li H. Zhang Y. Ji Z.Z. Knockdown of TMPRSS3 inhibits gastric cancer cell proliferation, invasion and EMT via regulation of the ERK1/2 and PI3K/Akt pathways Biomed. Pharmacother. 107 2018 841 848 30142546
54 Wang J.Y. Jin X. Li X.F. Knockdown of TMPRSS3, a transmembrane serine protease, inhibits proliferation, migration, and invasion in human nasopharyngeal carcinoma cells Oncol. Res. 26 2018 95 101 28409556
55 Guipponi M. Toh M.Y. Tan J. Park D. Hanson K. Ballana E. Kwong D. Cannon P.Z.F. Wu Q. Gout A. An integrated genetic and functional analysis of the role of type II transmembrane serine proteases (TMPRSSs) in hearing loss Hum. Mutat. 29 2008 130 141 17918732
56 Tang P.C. Alex A.L. Nie J. Lee J. Roth A.A. Booth K.T. Koehler K.R. Hashino E. Nelson R.F. Defective Tmprss3-associated hair cell degeneration in inner ear organoids Stem Cell Rep. 13 2019 147 162
57 Li Y. Peng A. Ge S. Wang Q. Liu J. miR-204 suppresses cochlear spiral ganglion neuron survival in vitro by targeting TMPRSS3 Hear. Res. 314 2014 60 64 24924414
58 Chen Y.S. Cabrera E. Tucker B.J. Shin T.J. Moawad J.V. Totten D.J. Booth K.T. Nelson R.F. TMPRSS3 expression is limited in spiral ganglion neurons: implication for successful cochlear implantation J. Med. Genet. 59 2022 1219 1226 35961784
59 Eppsteiner R.W. Shearer A.E. Hildebrand M.S. Deluca A.P. Ji H. Dunn C.C. Black-Ziegelbein E.A. Casavant T.L. Braun T.A. Scheetz T.E. Prediction of cochlear implant performance by genetic mutation: the spiral ganglion hypothesis Hear. Res. 292 2012 51 58 22975204
60 Holder J.T. Morrel W. Rivas A. Labadie R.F. Gifford R.H. Cochlear implantation and electric acoustic stimulation in children with TMPRSS3 genetic mutation Otol. Neurotol. 42 2021 396 401 33555745
61 Colella P. Ronzitti G. Mingozzi F. Emerging issues in AAV-mediated in vivo gene therapy Mol. Ther. Methods Clin. Dev. 8 2018 87 104 29326962
62 Tao Y. Huang M. Shu Y. Ruprecht A. Wang H. Tang Y. Vandenberghe L.H. Wang Q. Gao G. Kong W.J. Chen Z.Y. Delivery of adeno-associated virus vectors in adult mammalian inner-ear cell subtypes without auditory dysfunction Hum. Gene Ther. 29 2018 492 506 29130354
63 Suzuki J. Hashimoto K. Xiao R. Vandenberghe L.H. Liberman M.C. Cochlear gene therapy with ancestral AAV in adult mice: complete transduction of inner hair cells without cochlear dysfunction Sci. Rep. 7 2017 45524
64 Omichi R. Yoshimura H. Shibata S.B. Vandenberghe L.H. Smith R.J.H. Hair cell transduction efficiency of single- and dual-AAV serotypes in adult murine cochleae Mol. Ther. Methods Clin. Dev. 17 2020 1167 1177 32518805
65 Nass S.A. Mattingly M.A. Woodcock D.A. Burnham B.L. Ardinger J.A. Osmond S.E. Frederick A.M. Scaria A. Cheng S.H. O'Riordan C.R. Universal method for the purification of recombinant AAV vectors of differing serotypes Mol. Ther. Methods Clin. Dev. 9 2018 33 46 29349097
66 Noh B. Rim J.H. Gopalappa R. Lin H. Kim K.M. Kang M.J. Gee H.Y. Choi J.Y. Kim H.H. Jung J. In vivo outer hair cell gene editing ameliorates progressive hearing loss in dominant-negative Kcnq4 murine model Theranostics 12 2022 2465 2482 35265220
67 Shu Y. Tao Y. Wang Z. Tang Y. Li H. Dai P. Gao G. Chen Z.Y. Identification of adeno-associated viral vectors that target neonatal and adult mammalian inner ear cell subtypes Hum. Gene Ther. 27 2016 687 699 27342665
68 Sievers F. Wilm A. Dineen D. Gibson T.J. Karplus K. Li W. Lopez R. McWilliam H. Remmert M. Söding J. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega Mol. Syst. Biol. 7 2011 539 21988835
69 Shearer A.E. Tejani V.D. Brown C.J. Abbas P.J. Hansen M.R. Gantz B.J. Smith R.J.H. In vivo electrocochleography in hybrid cochlear implant users implicates TMPRSS3 in spiral ganglion function Sci. Rep. 8 2018 14165
70 Miyagawa M. Nishio S.Y. Sakurai Y. Hattori M. Tsukada K. Moteki H. Kojima H. Usami S.I. The patients associated with TMPRSS3 mutations are good candidates for electric acoustic stimulation Ann. Otol. Rhinol. Laryngol. 124 Suppl 1 2015 193S 204S 25770132
71 Müller M.K. Jovanovic S. Keine C. Radulovic T. Rübsamen R. Milenkovic I. Functional development of principal neurons in the anteroventral cochlear nucleus extends beyond hearing onset Front. Cell. Neurosci. 13 2019 119 30983974
72 Xue Y. Hu X. Wang D. Li D. Li Y. Wang F. Huang M. Gu X. Xu Z. Zhou J. Gene editing in a Myo6 semi-dominant mouse model rescues auditory function Mol. Ther. 30 2022 105 118 34174443
73 György B. Nist-Lund C. Pan B. Asai Y. Karavitaki K.D. Kleinstiver B.P. Garcia S.P. Zaborowski M.P. Solanes P. Spataro S. Allele-specific gene editing prevents deafness in a model of dominant progressive hearing loss Nat. Med. 25 2019 1123 1130 31270503
74 Xing Y. Samuvel D.J. Stevens S.M. Dubno J.R. Schulte B.A. Lang H. Age-related changes of myelin basic protein in mouse and human auditory nerve PLoS One 7 2012 e34500
75 Ivarsdottir E.V. Holm H. Benonisdottir S. Olafsdottir T. Sveinbjornsson G. Thorleifsson G. Eggertsson H.P. Halldorsson G.H. Hjorleifsson K.E. Melsted P. The genetic architecture of age-related hearing impairment revealed by genome-wide association analysis Commun. Biol. 4 2021 706 34108613
76 Ołdak M. Lechowicz U. Pollak A. Oziębło D. Skarżyński H. Overinterpretation of high throughput sequencing data in medical genetics: first evidence against TMPRSS3/GJB2 digenic inheritance of hearing loss J. Transl. Med. 17 2019 269 31412945
77 Leone M.P. Palumbo P. Ortore R. Castellana S. Palumbo O. Melchionda S. Palladino T. Stallone R. Mazza T. Cocchi R. Carella M. Putative TMPRSS3/GJB2 digenic inheritance of hearing loss detected by targeted resequencing Mol. Cell. Probes 33 2017 24 27 28263784
78 Sena-Esteves M. Gao G. Introducing genes into mammalian cells: viral vectors Cold Spring Harb. Protoc. 2020 2020 095513 10.1101/pdb.top095513
79 Liang S.Q. Walkey C.J. Martinez A.E. Su Q. Dickinson M.E. Wang D. Lagor W.R. Heaney J.D. Gao G. Xue W. AAV5 delivery of CRISPR-Cas9 supports effective genome editing in mouse lung airway Mol. Ther. 30 2022 238 243 34695545
80 Hu Z. Ulfendahl M. Olivius N.P. Central migration of neuronal tissue and embryonic stem cells following transplantation along the adult auditory nerve Brain Res. 1026 2004 68 73 15476698
81 Zhang L. Jiang H. Hu Z. Concentration-dependent effect of nerve growth factor on cell fate determination of neural progenitors Stem Cells Dev. 20 2011 1723 1731 21219132
